A multivariate and non-parametric batch effect correction framework based on Projection to Latent Structures Discriminant Analysis for microbiome data. This repository contains the R package hosted on Bioconductor.

Yiwen (Eva) Wang

(macOS users only: Ensure you have installed XQuartz first.)

Make sure you have the latest R version and the latest BiocManager package installed following these instructions.

## install BiocManager if not installed
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

Ensure the following returns TRUE, or follow the guidelines provided by the output.

BiocManager::valid()

You can install PLSDAbatch using the following code:

BiocManager::install('PLSDAbatch')

Install the GitHub version with:

# without vignette
BiocManager::install("EvaYiwenWang/PLSDAbatch") 

# with vignette
## Install CRAN packages for vignette
cran_pkgs <- c("pheatmap", "vegan")
to_install <- cran_pkgs[!cran_pkgs %in% installed.packages()[, "Package"]]
if (length(to_install) > 0) install.packages(to_install)

## Install Bioconductor packages for vignette
bioc_pkgs <- c("Biobase", "SummarizedExperiment")
to_install <- bioc_pkgs[!bioc_pkgs %in% installed.packages()[, "Package"]]
if (length(to_install) > 0) BiocManager::install(to_install)

devtools::install_github("https://github.com/EvaYiwenWang/PLSDAbatch", build_vignettes = T)

library(PLSDAbatch)

## names
ls('package:PLSDAbatch')

## names and details
lsf.str('package:PLSDAbatch')

browseVignettes("PLSDAbatch")

Open a new issue.

Wang, Y., & Lê Cao, K. A. (2023). PLSDA-batch: a multivariate framework to correct for batch effects in microbiome data. Briefings in Bioinformatics, 24(2), bbac622.

https://academic.oup.com/bib/article/24/2/bbac622/6991121 (The mentioned simulations and analyses in the paper are separately stored here.)

• submitted to Bioconductor.

• fixed bugs: the clash of functions from dependencies.

• Added a mode argument to PLSDA_batch().
• Added a criterion argument to linear_regres() to select P-values from the optimal model based on the specified criterion.
• Added a return.model arugument to linear_regres() to reduce memory usage when set to FALSE.
• Extended Scatter_Density() to support any multivariate method that returns component scores, including PCA and PLS, with corresponding arguments updated.
• Added lighten() and darken() functions for enhanced color generation.
• Refined multiple functions to improve usability.
• Updated the vignette accordingly.